Download Contingent fire effects on granivore removal of exotic woody plant

Biol Invasions
DOI 10.1007/s10530-013-0557-1
ORIGINAL PAPER
Contingent fire effects on granivore removal of exotic woody
plant seeds in longleaf pine savannas
Jacqueline S. Krall • Matthew G. Hohmann
Jennifer M. Fraterrigo
•
Received: 10 February 2013 / Accepted: 18 September 2013
Ó Springer Science+Business Media Dordrecht 2013
Abstract Prescribed fire is increasingly used to
inhibit woody encroachment into fire-dependent ecosystems, yet its effects on other processes influencing
invasion are poorly understood. We investigated how
fire influences exotic woody invasion through its
effects on granivore activity, and whether these effects
depend on the habitat in which seed predation takes
place. We quantified seed removal for four species of
exotic woody plants (Albizia julibrissin, Elaeagnus
umbellata, Melia azedarach and Triadica sebifera) in
17 sites in longleaf pine savanna that varied in time
since fire (one or three growing seasons post-fire) in the
sandhills region of North Carolina, USA. Within each
site, we established paired plots in upland and uplandwetland ecotone communities and presented seeds in
depots that allowed either arthropod, or arthropod and
small vertebrate access. We found that differences in
seed removal with time since fire were contingent on
habitat and granivore community. In ecotones, three of
four species had higher proportions of seeds removed
from plots that were three growing seasons post-fire
than plots one growing season post-fire, whereas only
T. sebifera showed this pattern in upland habitat.
Allowing vertebrate granivores access to seeds
enhanced seed removal, and this effect was strongest
in ecotone habitat. While granivores removed seeds of
all four plant species, removal of E. umbellata was
significantly higher than that of the other species,
suggesting that granivores exhibited seed selection.
These findings suggest that ecotone habitats in this
system experience greater seed removal than upland
habitats, particularly as time since fire increases, and
differences are mainly due to the activity of vertebrate
granivores. Such differences in seed removal, together
with seed selection, may contribute to variation in
exotic woody invasion of longleaf pine savannas.
Keywords Post-dispersal seed predation Woody encroachment Prescribed fire Rodents
Introduction
J. S. Krall J. M. Fraterrigo (&)
Department of Natural Resources and Environmental
Sciences, University of Illinois at Urbana–Champaign,
1102 S. Goodwin Ave., MC-047, Urbana, IL 61801, USA
e-mail: [email protected]
M. G. Hohmann
Construction Engineering Research Laboratory, US Army
Engineer Research and Development Center, Champaign,
IL 61826, USA
Invasions by exotic woody plant species can be
detrimental to native plant communities and ecosystems sensitive to changes in vegetation structure, such
as those dependent on frequent fire (Mandle et al.
2011). Savannas, for example, require fire to maintain
a sparse canopy cover and continuous grass understory, conditions which in turn promote fire (Bond and
123
J. S. Krall et al.
Keeley 2005; Beckage et al. 2009). Encroachment by
woody species, including exotics, can disrupt this
grass-fire cycle and alter the successional trajectory of
the system (Brooks et al. 2004; Mandle et al. 2011). In
the absence of a natural fire regime, prescribed fire is
increasingly used to inhibit tree and shrub encroachment into fire-dependent ecosystems (Sousa 1984),
but its effects on exotic woody invasion remain poorly
understood. To date, the majority of studies investigating fire effects on exotic invasion have focused on
‘‘bottom-up’’ control; that is, the effects of post-fire
conditions on plant performance and competition.
Such studies have yielded varied results. For example,
research shows that fire can reduce invasive populations via seed and plant mortality (Lonsdale and Miller
1993), and reductions in fecundity (Stevens and
Beckage 2010), yet may alternatively promote invasion of woody species by reducing native vegetation
cover (Kuppinger et al. 2010), and increasing germination (Ne’eman et al. 2004; Todorović et al. 2010).
Much less is known about how fire influences topdown processes affecting invasion, such as granivory.
Post-dispersal granivory is recognized as a key
process influencing woody plant recruitment (Hulme
1998). Recent studies from diverse ecosystems show
that granivores can affect the pattern and rate of woody
plant recruitment by altering the distribution and
viability of seeds (Schnurr et al. 2004; Côté et al. 2005;
Zwolak et al. 2010; Vaz Ferreira et al. 2011).
Granivory is expected to influence woody encroachment when seed limitation has a greater effect on
population growth rates than establishment limitation
(Clark et al. 2007). However, these constraints are not
mutually exclusive and are expected to vary through
space and time due to variation in fecundity, dispersal
and suitability of microsites, thereby influencing
invasion rates and patterns across the landscape
(Nathan and Muller-Landau 2000; Satterthwaite
2007). For instance, strong seed limitation is anticipated at invasion boundaries during local range
expansion because propagule pressure is often limited
by dispersal capacity, low population densities, and
Allee effects (Taylor and Hastings 2005; Theoharides
and Dukes 2007).
Small mammals are important granivores of woody
plant seeds, potentially more so than invertebrate
granivores due to size related constraints on seed
handling for arthropods and the positive relationship
between plant height and seed size (Thompson and
123
Rabinowitz 1989; Leishman et al. 2000). Seed
removal rates are correlated with small mammals
preferences for specific habitat characteristics (Hulme
1994), including high levels of vegetative cover
(Kotler 1984; Manson and Stiles 1998) and downed
woody debris for predator refuge (Schnurr et al. 2004).
Fire may alter granivore activity by modifying habitat
characteristics. For instance, fire can reduce vegetation cover, which may increase the perceived predation risk of small mammals and change their foraging
patterns (Jacob and Brown 2000; Orrock et al. 2004).
Fire can also reduce litter depth (Bond and Keeley
2005), increasing or decreasing small mammal and
arthropod foraging activity (Tester 1965; Krefting and
Ahlgren 1974; Fewell 1988; Reed et al. 2006; Gibb
and Parr 2010), and leading to varied levels of seed
removal.
The influence of fire on granivore activity is likely
to vary across space, however, because fire both
generates and responds to spatial heterogeneity in
vegetation structure at multiple scales. In managed
landscapes, prescribed fire is often applied to distinct
land units, which results in broad scale spatial
variation in length of time since fire. Additionally,
because these units may encompass different plant
community types that vary in underlying soil moisture
conditions, there may be differences in fire intensity
and vegetation recovery rate that increase spatial
heterogeneity at a finer scale (Kirkman et al. 2004).
Collectively, fire-induced spatial heterogeneity may
lead to heterogeneous seed removal and contribute to
between-habitat and landscape scale variation in
exotic woody plant invasions.
Within heterogeneous landscapes created by spatial
variation in plant communities and time since fire,
seed selection by granivores may also contribute to
differences in seed removal among plant species. Seed
removal may be influenced by morphological (Lundgren and Rosentrater 2007), chemical defense (Kollmann et al. 1998), and nutritional traits (Kelrick et al.
1986). For example, small seeds are preferentially
chosen by arthropods due to weight and size limitations on movement (Rey et al. 2002), whereas large
seeds are more likely to be consumed by small
mammals (Brown and Heske 1990; Matias et al.
2009). Seed selection may also depend on how habitat
structure changes with time since fire because seed
preference can be influenced by perceived predation
risk (Bowers 1990).
Exotic seed removal and fire in longleaf pine savanna
In this study, we investigate how fire influences
exotic woody plant species through its effects on
granivore activity. We focus on landscape-scale
variation associated with differences in time since fire
and between-habitat-scale variation due to differential
effects of fire within a management unit. Our specific
objectives were to: (1) quantify patterns of seed
removal between upland savanna and upland-wetland
ecotone habitats, which represent low and high
vegetation structure, respectively, (2) quantify patterns of seed removal between upland savanna and
upland-wetland ecotone habitats that were burned
either one or three growing seasons prior, (3) compare
seed removal of four exotic woody plant species that
vary in seeds traits; and (4) assess the importance of
granivore community type (i.e., arthropod vs. small
mammal) for explaining variation in removal. We
expected seed removal to increase with vegetation
structure and therefore increase with time since fire
presumably due to small mammal predator avoidance
behavior. We further expected habitat type to modulate this effect because fire intensity and vegetation
recovery vary with habitat. We expected that species
would be differentially selected by arthropod and
small mammal granivores potentially due to differences in seed characteristics. By assessing the magnitude of post-dispersal exotic woody seed removal in
a savanna ecosystem under threat of woody encroachment, additional insights can be gained about landscape-scale vulnerability to exotic plant invasions and
focused efforts can be made to help manage emerging
invasions.
Methods
Study area and site selection
The study was carried out on Fort Bragg Army
installation near Fayetteville, North Carolina
(35°070 N, 79°100 W). The installation covers an area
of 73,468 ha and ranges in elevation from 43 to
176 m. Mean temperature is 26 °C during the summer
and 7 °C during the winter, and mean annual rainfall
in the region is 120 cm with September–November
and March–April being the driest months (Sorrie et al.
2006). The installation is located in the Fall-line
Sandhills region near the northern extent of the former
longleaf pine–wiregrass savanna ecosystem and is
comprised of multiple plant community types that
separate along topographic, soil, and fire gradients.
Natural upland savanna sites on the installation are
dominated by longleaf pine (Pinus palustris) and a
sparse understory of wiregrass (Aristida stricta),
legumes, scrub oaks, ericaceous shrubs, and xerophytic forbs. The ecotonal zone between upland
savanna and wetland pocosins (hereafter ‘‘ecotone’’)
are characterized by an overstory of longleaf pine, red
maple (Acer rubrum), and tulip tree (Liriodendron
tulipifera), as well as a dense understory that includes
cane (Arundinaria tecta), ferns (e.g., Osmunda cinnamomea, Pteridium aquilinum), and woody shrubs
(e.g., Clethra alnifolia, Ilex glabra, Lyonia lucida)
(Sorrie et al. 2006). Small mammal survey data on Fort
Bragg indicate local mammalian granivores include:
Glaucomys volans, Microtus pinetorum, Mus musculus, Peromyscus gossypinus, Peromyscus leucopus,
Reithrodontonmys humulis, Sciurus carolinensis, Scirus niger, Sigmodon hispidus, and Sylvilagus palustris
(JM McCallister, unpubl. data). Arthropod granivores
include: ground beetles (Coleoptera; Carabidae),
harvester ants (Hymenoptera; Formicidae), and crickets (Orthoptera; Gryllidae) (M. Hohmann, personal
obs.). We did not perform a separate census of the
granivore community for this study because we were
mainly interested in discerning general trends in
removal by different granivore groups as a function
of habitat type and time since fire.
Fort Bragg is under a managed fire cycle in
which growing (April–July) and dormant (January–
March) season fires are applied to 4.0–81.0 ha land
management units locally called burn blocks
approximately every 3–5 years. We selected nine
burn blocks that were three growing seasons postfire (last burned 2009), and seven burn blocks that
were one growing season post-fire (last burned
during the 2011 dormant season). Within each burn
block we established a pair of 10 m 9 40 m
(0.04 ha) plots, one in upland and one in ecotone
habitat. A second pair of plots was established in
one of the largest recently burned blocks. All plots
were separated by at least 115 m, a distance larger
than the average home range of the most common
small vertebrate seed predator, P. leucopus (32 m
radius) (Lackey et al. 1985). We did not establish
controls in unburned sites because such sites no
longer supported plant communities characteristic of
the savanna ecosystem.
123
J. S. Krall et al.
Focal species
Field measurements
We focused on four deciduous woody exotic species:
Albizia julibrissin, Elaeagnus umbellata, Melia azedarach and Triadica sebifera. All of these species are
considered invasive in the southeastern USA, but are
currently rare across Fort Bragg and surrounding
counties. All four species have the potential to
establish and bear fruit under the fire-management
cycle on Fort Bragg, as well as re-sprout after fireinduced stem death (i.e., top-kill) (Munger 2003;
Waggy 2009; Meyer 2010, 2011). Fruits mature
locally between July and November (Weakley
2012). Seeds are primarily avian dispersed, except
for A. julibrissin, which largely disperses via wind.
Individuals of each species may produce thousands of
viable seeds per growing season, and except for E.
umbellata, have seed banks of short to moderate
longevity. Disturbance promotes the establishment of
all four species, which prefer full sun, but tolerate
shade or partial shade.
Seeds of each species were bulk-collected from
populations on or near Fort Bragg from September to
November 2011 and processed to mimic the state in
which they occur following primary dispersal. For M.
azedarach and E. umbellata, we manually removed
the fleshy coatings of fruits, rinsed off the pulp and
allowed the seeds to dry for 48 h. Albizia julibrissin
and T. sebifera seeds were sorted from attached pods
and capsules and allowed to dry for 24 h. Melia
azedarach, E. umbellata and T. sebifera seeds were
presented still enclosed in their protective endocarp.
To characterize average seed size, we measured dry
seed mass, length, and width of 100 randomly chosen
seeds of each species prior to storage in paper bags at
room temperature (Table 1).
We measured substrate and vegetation characteristics
within each of the 34 study plots during November and
December 2011. The percent cover of ground layer
understory vegetation\ 2.0 m was visually estimated
within fourteen 1 m2 quadrats, one every 5 m along
two parallel 40 m transects separated by 10 m. We
measured litter depth at 10 points within a 1 m radius
of the 5, 15, 25, and 35 m points along a 40 m transect
through the plot center.
We conducted two 16-days seed removal presentations from November 5–21 and December 5–21,
2011. These time periods were chosen to coincide with
the natural dispersal timing of the focal species.
Multiple presentation periods allowed us to determine
if removal patterns remained consistent within a single
season (Boman and Casper 1995; Orrock and Damschen 2005). Presentations coincided with the timing of
natural seed dispersal of the four focal species. To
control for the influence of moonlight on vertebrate
foraging behavior, presentations were timed to occur
during similar lunar phases of waxing gibbous through
waning crescent (Orrock et al. 2004). Within each plot
we installed a pair of seed removal depots 1 m apart at
5, 15, 25, and 35 m along a transect that bisected the
plot (n = 8 depots per plot; n = 272 depots total).
Empty depots were placed in the field approximately
3 months prior to seed presentation periods. Two
depot types were used: (1) accessible only to arthropods, and (2) accessible to arthropods and small
mammals. All depots were inverted translucent plastic
buckets 12.7 cm in height and 21.9 cm in diameter
with two 15 cm 9 8 cm openings on opposite sides,
and the base removed. Mesh hardware cloth (1.6 cm2
gauge) was used to bar access via the inverted base of
Table 1 Seed removal (mean ± SE) by presentation period and seed size and weight for four exotic woody species presented in
seed removal depots. Species are listed in order of increasing size
Species
Elaeagnus umbellata
Seed Removal (%)
Length (mm)b
Weight (g)a
November
December
24.7 ± 2.2
19.5 ± 2.3
4.0–7.0 (5.5 ± 0.07)
1.5–2.0 (2.0 ± \0.01)
0.014 (±0.002)
4.4 ± 0.6
3.8 ± 0.6
4.0–9.0 (6.7 ± 0.10)
2.0–5.0 (3.1 ± 0.05)
0.037 (±0.007)
Albizia julibrissin
Triadica sebifera
3.8 ± 0.6
4.2 ± 0.7
5.0–10.0 (7.2 ± 0.09)
Melia azedarach
2.2 ± 0.5
1.4 ± 0.4
5.0–12.0 (9.0 ± 0.14)
a
Values are: mean (±SD)
b
Values are: minimum–maximum (mean ± SD)
123
Width (mm)b
3.0–8.0 (5.3 ± 0.08))
5.0–11.0 (8.4 ± 0.13)
0.116 (±0.03)
0.395 (±0.10)
Exotic seed removal and fire in longleaf pine savanna
Vegetation and substrate measurements were analyzed
separately using a linear mixed model, with time since
fire, habitat, and time since fire 9 habitat as fixed
effects and burn block nested within time since fire as a
random effect (PROC MIXED, SAS v.9.3, Cary, NC,
USA). We examined simple main effects to evaluate
the influence of specific levels of each factor within
interactions using the Slice statement.
To analyze the main effects and interactions of
habitat type, time since fire, depot type, and species on
post-dispersal seed removal, we fit a generalized linear
mixed model to percentage seed removal using the
PROC GLIMMIX procedure (SAS v.9.3, Cary, NC,
USA). This procedure allows for the analysis of mixed
models with non-normal response variables by
modeling specific distribution structures (Bolker
et al. 2009). We initially compared the same model
structure with two distributions for the response
variable, the binomial and Poisson, to determine
which distribution best fit the data (Johnson and
Omland 2004). A binomial distribution and logit link
function, which utilizes an ‘‘event per trial’’ response
variable, yielded the lowest AICc value and was
selected. We analyzed presentation periods separately
using individual linear mixed models with habitat
type, time since fire, depot type, species, and all
interactions as fixed effects. Random variables
included terms for plots and depot pairs to represent
the sampling structure with nesting of random variables (Onofri et al. 2010). Following model selection,
we compared multiple combinations of random statements using the covariance parameter estimates to
Results
Vegetation structure
Understory cover differed significantly with habitat type
(Fig. 1). Cover was 57 % higher in ecotone than upland
habitat, averaging 55 ± 3 SE % and 35 ± 4 %, resp.
(F1,13.9 = 14.9, P = 0.002). There was a trend for
higher understory vegetation cover in ecotone habitats
Understory vegetation cover (%)
Statistical analysis
determine the random variable structure that minimized redundancy and accounted for the most variation in the data. We tested for differences in seed
removal among different levels of fixed effects using
least square means tests and a Tukey’s correction for
multiple comparisons. Simple main effects were
examined with the Slice statement in the PROC
GLIMMIX procedure.
70
(A)
60
One growing season post-fire
Three growing seasons post-fire
50
40
30
20
10
0
4
Litter depth (cm)
all depots, and the side openings of the ‘‘accessible
only to arthropod’’ depots. Similarly designed depots
have been used successfully in other seed removal
studies (Mattos and Orrock 2010; Craig et al. 2011).
The attached bucket lid of each depot had a  lip to
control for incidental seed loss from natural elements
such as wind or rain. Depots were secured in place
with four steel landscape pins. For both seed presentations we covered the attached bucket lid of each
depot with sand and deposited 10 seeds of each of the
four species on the surface. After 15 d, we collected
the sand and seeds from each depot, sifted the
contents, and counted the remaining intact and damaged seeds.
3
(B)
*
*
2
1
0
Upland
Ecotone
Habitat type
Fig. 1 Mean (?SE) understory vegetation cover (a) and litter
depth (b) in upland and ecotone habitats one or three growing
seasons post-fire. The means and standard errors displayed here
are not corrected for the nested design of the study. Asterisks
represent statistically significant differences within habitat types
between time post-fire groups (*P B 0.05)
123
J. S. Krall et al.
Habitat type, species and depot type had strong
main effects on seed removal during both presentation
periods (Table 2). Seed removal was 77 % higher in
ecotone than upland habitat [15.2 ± 1.1 % (SE) and
8.6 ± 0.7 %, resp.] in November, and 201 % higher in
ecotone than upland habitat (20.3 ± 1.3 and
6.8 ± 0.6 %, resp.) in December. Among species,
mean seed removal ranged from 6.4 to 27.6 % in
November and 8.5 to 24.3 % in December. In both
periods, removal of E. umbellata, the smallest seeded
species, was more than twofold higher than that of
other species (Table 1; Fig. 2). Seed removal was also
consistently higher from depots that allowed both
arthropods and small mammals access than from those
that allowed only arthropods [November: 14.9 ±
1.1 % (SE) and 9.0 ± 0.8 %, resp.; December:
21.2 ± 1.3 and 5.8 ± 0.6 %, resp.].
There were also significant three-way interactions
among these factors in both periods (Table 2). In
November, seed removal from arthropod only depots
was higher in ecotone than upland habitat for E.
umbellata (F1,70.98 = 13.8, P \ 0.001), and similar
for all other species (M. azedarach and A. julibrissin:
P [ 0.14; T. sebifera: F1,418.2 = 3.07, P = 0.08)
with time since fire, but this difference was not
significant (Fig. 1a).
Litter depth also varied with habitat type. On
average, litter depth was 50 % higher in ecotone
than upland habitat (mean ± SE; 2.1 ± 0.3 and
1.4 ± 0.3 %, respectively) (F1,27.5 = 5.6, P = 0.03).
Time since fire had a significant effect on litter depth,
which was on average 253 % higher in plots three
growing seasons post-fire than in plots one growing
season post-fire (2.7 ± 0.3 cm and 0.8 ± 0.1 cm,
resp.) (F1,27.5 = 41.4, P \ 0.001). The effect of time
since fire on litter depth did not vary with habitat type
(Fig. 1b).
Seed removal
Overall, the number of depots experiencing seed
removal was similar between presentation periods,
with at least one seed removed from 246 depots in
November and 222 depots in December. Average
seed removal was 8.8 % in November and 8.7 % in
December. Several depots were damaged (six in
November and one in December) making seeds
irretrievable and counts impossible.
Table 2 ANOVA results testing for significance of seed removal by main effects and interactions during November and December
seed presentations
November seed presentation
Num df
Den df
F
Time since fire
1
41.5
0.0
Habitat
1
41.5
7.15
December seed presentation
P
Num df
Den df
F
P
0.96
1
35.3
0.011
1
35.3
16.2
149.2
<0.001
84.8
<0.001
Main effects
Species
3
1,032
Depot type
1
447
229.
33.4
<0.001
3
1,051
<0.001
1
481
0.23
0.64
<0.001
Interactions
Time since fire 9 habitat
1
41.5
0.1
0.76
1
35.3
1.34
0.25
Time since fire 9 species
3
1,032
0.62
0.60
3
1,051
4.62
0.003
Time since fire 9 depot type
1
447
3.86
0.050
1
481
0.03
0.86
Habitat 9 species
3
1,032
7.75
\0.001
3
1,051
3.34
0.019
Habitat 9 depot type
1
447
Species 9 depot type
Time since fire 9 habitat 9 species
3
3
1,032
1,032
3.90
0.049
1
481
22.9
<0.001
15.0
0.02
<0.001
0.99
3
3
1,051
1,051
14.1
3.53
<0.001
0.014
Time since fire 9 habitat 9 depot type
1
447
1.49
0.22
1
481
0.37
0.55
Time since fire 9 species 9 depot type
3
1,032
3.18
0.023
3
1,051
3.52
0.015
Habitat 9 species 9 depot type
3
1,032
<0.001
3
1,051
6.78
<0.001
Time since fire 9 habitat 9 species 9 depot type
3
1,032
0.083
3
1,051
1.54
0.20
Values in bold are significant at alpha B 0.05
123
10.1
2.23
Exotic seed removal and fire in longleaf pine savanna
60
(B) Arthropods and small mammals
(A) Arthropods
Seeds removed (%)
50
Upland
Ecotone
**
40
30
**
**
20
10
0
60
(D) Arthropods and small mammals
(C) Arthropods
**
Seeds removed (%)
50
**
**
40
**
30
20
10
0
h
a
eli
M
bif
ica
a
iad
Tr
in
a
er
ac
ar
d
ze
ta
iss
ia
i
jul
nu
um
g
ea
a
El
ar
be
s
biz
Al
h
a
eli
M
bif
d
ze
ica
a
iad
Tr
in
a
er
ac
lla
br
se
ta
iss
lla
br
se
ia
i
jul
biz
be
us
um
gn
Al
a
lae
E
Fig. 2 Mean (?SE) seed removal by species, from arthropod
only (a, c) and arthropod and small mammal (b, d) depots placed
in ecotone and upland habitats for November (top panels) and
December (bottom panels) presentations. The means and
standard errors displayed here are not corrected for the nested
design of the study. Individual species are ordered by decreasing
size. Asterisks represent statistically significant differences
within species between habitat types, by depot type
(*P B 0.05, **P B 0.01)
(Fig. 2). In contrast, all species showed a trend for
higher seed removal in ecotone than upland habitat
when seeds were placed in depots that allowed access
to both arthropods and small mammals. Differences
were significant for M. azedarach and A. julibrissin
(for both, P \ 0.001), but not for E. umbellata
(F1,70.25 = 2.83, P = 0.09) and T. sebifera
(F1,130.6 = 0.85, P = 0.36) (Fig. 2) Species-level patterns in seed removal in December were comparable to
those of November. Removal from arthropod only
depots was higher in ecotone than upland habitat for E.
umbellata (F1,61.52 = 2.62, P = 0.11) and similar
between habitat types for all other species
(P [ 0.31). Seed removal from depots that allowed
small mammals access was higher in ecotone than
upland habitat for all species (P \ 0.001) (Fig. 2).
Seed removal did not vary consistently with time
since fire during either presentation period (Table 2).
Mean seed removal in areas one and three growing
seasons post-fire was 12.2 ± 1.0 % (SE) and
11.6 ± 0.9 % in November and 11.8 ± 1.0 and
15.1 ± 1.1 % in December. There were, however,
significant two- and three-way interactions with time
since fire (Table 2), indicating that fire effects on seed
removal were contingent on other factors.
Habitat influenced the effects of time since fire
primarily in December (Table 2). In this period, seed
removal in upland habitat was 402 % higher in plots one
123
J. S. Krall et al.
123
25
Seeds removed (%)
(A) Upland
One growing season post-fire
Three growing seasons post-fire
20
15
10
*
5
0
40
(B) Ecotone
Seeds removed (%)
growing season post-fire than three growing seasons
post-fire for M. azedarach [2.8 ± 1.4 % (SE) and
0.6 ± 0.3 %, resp.; F1,440.1 = 3.59, P = 0.05], but
similar for all other species (P [ 0.16) (Fig. 3). We
observed the opposite pattern in ecotone habitat, such
that seed removal was 41 % lower in areas one growing
season post-fire than those three growing seasons postfire for M. azedarach (11.9 ± 3.6 and 20.0 ± 4.2 %,
resp.; F1,240.8 = 4.18, P = 0.04) and 47 % lower for T.
sebifera (14.1 ± 3.7 % and 26.8 ± 4.0 %, resp.;
F1,52.14 = 3.23, P = 0.07) (Fig. 3). Seed removal for
the other species was similar with time since fire in
ecotone habitat (P [ 0.43) (Fig. 3).
Depot type also influenced the effect of time since
fire on seed removal for M. azedarach and T. sebifera.
Removal of M. azedarach seeds from depots allowing
access to both arthropods and small mammals was
significantly higher in November in plots one growing
season post-fire than in plots three growing seasons
post-fire [15.2 ± 3.5 % (SE) and 8.3 ± 2.7 %, resp.;
F1,196.9 = 10.7, P = 0.001; data not shown]. In contrast, removal of T. sebifera was significantly lower in
December in plots one growing season post-fire than in
plots three growing seasons post-fire (14.6 ± 3.7 and
29.3 ± 3.9 %, resp.; F1,84.87 = 11.1, P = 0.001; data
not shown). Removal of all other species was similar
between fire treatments within depot types (P [ 0.13).
Additionally, habitat and depot type jointly influenced the effect of time since fire on removal of M.
azedarach and T. sebifera seeds (Fig. 4). In ecotone
habitat in December, seed removal from depots
allowing both arthropod and small mammal access
was 38 and 49 % lower in plots one growing season
post-fire than in plots three growing seasons post-fire
for M. azedarach (F1,58.3 = 4.99, P = 0.03) and T.
sebifera (F1,56.4 = 10.2, P = 0.002), respectively. In
upland habitat, however, removal of M. azedarach
seeds from arthropod and small mammal depots was
398 and 500 % higher in plots one growing season
post-fire in November (F1,344 = 9.63, P = 0.002) and
December (F1,395 = 5.14, P = 0.02), respectively
(Fig. 4). There was also a trend in December for
lower seed removal for T. sebifera in upland plots one
growing season post-fire than in plots three growing
seasons post-fire (Fig. 4), but the difference was not
significant (F1,56.4 = 2.80, P = 0.09). For A. julibrissin and E. umbellata, seed removal was similar one
and three growing seasons post-fire within each
combination of habitat and depot type.
30
*
20
10
0
ch
a
eli
az
M
ica
iad
Tr
ta
sin
ra
ife
ra
a
ed
lla
ris
b
se
ia
be
ib
jul
s
nu
biz
um
ag
Al
lae
E
Fig. 3 Mean (?SE) seed removal in December by species,
from seed depots located in plots one or three growing seasons
post-fire in upland (a) and ecotone (b) habitats. The means and
standard errors displayed here are not corrected for the nested
design of the study. Individual species are ordered by decreasing
seed size. Asterisks represent statistically significant differences
within species, between plots one or three growing seasons postfire, by habitat type (*P B 0.05)
Discussion
In fire-dependent ecosystems, recent fire history may
influence patterns of exotic woody invasion by
altering granivore activity. Few studies, however,
have examined how fire effects vary spatially due to
heterogeneity in time since fire or vegetation structure,
or have evaluated the factors that modulate these
relationships. We found that habitat type (defined by
plant community), exotic species, and granivore
community strongly affected seed removal, as well
as influenced the relationship between time since fire
and seed removal in longleaf pine savannas. For three
of four species, seed removal was highest in ecotones
three growing seasons post-fire. Seed depots that
Exotic seed removal and fire in longleaf pine savanna
60
50
Seeds removed (%)
(B) Ecotone
(A) Upland
One growing season post-fire
Three growing seasons post-fire
40
30
20
**
10
0
60
(C) Upland
50
Seeds removed (%)
(D) Ecotone
**
*
40
30
20
*
10
0
ch
ia
el
M
az
ca
di
ia
Tr
sin
ra
ife
ra
a
ed
zia
j
s
nu
a
ed
um
g
ea
la
E
ia
el
M
az
ca
di
ia
Tr
sin
ra
ife
ra
be
ib
ul
bi
Al
ch
ta
lla
ris
b
se
zia
be
ib
ul
j
s
nu
bi
Al
ta
lla
ris
b
se
um
ag
e
la
E
Fig. 4 Mean (?SE) seed removal by species from seed depots
that allowed access to both arthropod and small mammals and
were placed in plots one or three growing seasons post-fire in
upland (a, c) and ecotone (b, d) habitats for November (top
panels) and December (bottom panels) presentations. The
means and standard errors displayed here are not corrected for
the nested design of the study. Individual species are ordered by
decreasing seed size. Asterisks represent statistically significant
differences within species between plots one or three growing
seasons post-fire, by habitat type (*P B 0.05, **P B 0.01)
allowed small mammals and arthropods access
showed more than twofold higher removal than depots
that only allowed arthropods access, indicating that
small mammals were more effective at removing
seeds of the focal species. These findings suggest that
ecotone habitat supporting small mammal granivores
experiences greater post-dispersal seed removal than
upland habitats; and, for at least two species, postdispersal seed removal increases in ecotone habitat as
time since fire increases and vegetation recovers.
Fire both drives and responds to spatial and
temporal variation in vegetation structure within the
longleaf pine ecosystem (Kirkman et al. 2004).
Results from this study support this idea, and suggest
that differences in vegetation structure associated
with habitat type and length of time since fire
contribute to spatial heterogeneity at between-habitat
and landscape scales. Vegetative cover was higher in
ecotone than upland habitat; and for ecotone habitat,
time played a role in recovery, with greater percentage cover three growing seasons post-fire than one.
Previous studies show that vegetation recovery
following fire in the longleaf ecosystem is affected
by fire behavior (Thaxton and Platt 2006; Ellair and
Platt 2012), microsite conditions (Olson and Platt
1995; Drewa et al. 2002), component species’
morphological and physiological traits (Hiers and
Mitchell 2007; Romero et al. 2009), and plant size at
time of burn (Rebertus et al. 1989; Grady and
Hoffmann 2012).
123
J. S. Krall et al.
Seed removal patterns strongly reflected habitatassociated differences in post-fire vegetation cover
and recovery. For example, we found that overall seed
removal was higher in ecotone than upland habitat
regardless of time since fire. Additionally, for three of
the four species studied, seed removal was highest in
ecotone habitat three growing seasons post-fire. While
this difference was only significant for two species, it
indicates that even the short 3-year interval between
fires in the study system can create ecologically
significant differences in habitat structure that influence small mammal granivore activity. These results
are consistent with those of other studies showing
higher seed removal by mammalian granivores with
greater vegetation cover (e.g., Mittelbach and Gross
1984; Hulme 1994; Alcantara et al. 2000), and are
presumably due to the decreasing perceived predation
risk of granivores with increasing vegetation cover
(Bowers 1993; Garcia-Castano et al. 2006; Pons and
Pausas 2007). However, they are in contrast to other
studies that have shown seed removal increases after
fire in both grasslands (Reed et al. 2004) and forests
(Côté et al. 2005; Zwolak et al. 2010), due to increased
granivore foraging efficiency (Reed et al. 2005;
Zwolak et al. 2012). Furthermore, the opposing effects
of time since fire on removal of M. azedarach and T.
sebifera seeds between habitats and presentation
periods emphasize the importance of examining
potential spatial, temporal and species-specific variation in seed removal. Ecotone habitats may be at
greater risk to exotic woody plant invasion directly
after fire not only due to reduced seed removal, but
also because the consumption of vegetation and litter
by fire can increase the availability of microsites for
germination (e.g., Kuppinger et al. 2010; Mazia et al.
2010).
Although granivores removed seeds of all four
woody exotic species, the smallest seeded species, E.
umbellata, was both strongly selected and the only
species for which direct evidence of in situ seed
predation was found, in the form of empty endocarps.
Using seeds of a similar weight and size as this
experiment, others have found similar selective
removal of smaller seeds by small mammals (Alcantara et al. 2000), and both arthropods and small
mammals (Carrillo-Gavilan et al. 2010). However,
as seed size did not consistently predict removal rates,
other seed traits may have influenced removal. For
example, seed strength, impacted by seed thickness,
123
has been shown to affect seed selection in granivores
(Lundgren and Rosentrater 2007). Seed chemical
defenses also vary among plant species to deter
vertebrate granivores, which respond strongly to
visual, gustatory olfactory signals (Janzen 1971).
Seeds of several Albizia species and many other
legumes contain chemicals toxic to rodent granivores
(Duncan et al. 1953; Grant et al. 1995), and M.
azedarach fruits contain toxic tetranortriterpenes
(Oelrichs et al. 1983). Indeed, M. azedarach and A.
julibrissin were removed less often from depots that
allowed small vertebrates access than the other
species, which have no known chemical defenses.
Nutritional differences among seeds may also be
important in influencing granivore foraging behavior
(Kelrick et al. 1986). Triadica sebifera seeds have
high levels of short- and long-chain fatty acids
(Terigar et al. 2010), and were primarily removed
when small vertebrates were allowed access, rather
than arthropods only.
Arthropod granivores removed relatively few seeds
of the focal species during the November and
December seed presentation periods. With colder
average temperatures during the night (9.0 and
11.2 °C, resp.) and day (11.8 and 14.5 °C, resp.), the
low removal rates could have been due to restrictions
in arthropod activity (Golley and Gentry 1964;
Niehaus et al. 2012). Seed removal from depots
allowing only arthropod access did not differ by
habitat type or time since fire, which suggests
understory vegetation cover and litter depth had no
effect on arthropod foraging. This is inconsistent with
other studies which show arthropod seed predation is
higher on bare ground than litter covered substrates
(Crist and Wiens 1994) and that fire increases
arthropod foraging distances (McCoy and Kaiser
1990) and seed predation (Côté et al. 2005). However,
we may not have been able to detect any effects of
habitat or time since fire due to low removal rates.
When seed removal could be unequivocally attributed
to arthropod granivores, E. umbellata seeds were
selected. Arthropods are more restricted in seed
transport capabilities than small vertebrates, and at
0.014 g, E. umbellata seeds are likely heavy for
arthropod removal (MacMahon et al. 2000; Lundgren
and Rosentrater 2007).
Our results suggest that small vertebrates are the
primary post-dispersal granivores of the focal species
during the fall across the longleaf pine ecosystem
Exotic seed removal and fire in longleaf pine savanna
studied. This is consistent with evidence suggesting
that vertebrates are the primary granivores in ecosystems regularly experiencing fire (Auld and Denham
2001; Reed et al. 2004), and of woody species having
comparably sized seeds (Ostfeld et al. 1997; Whelan
et al. 1991). The depot design limits inference to small
mammals, presumably rodents, and while birds were
not explicitly excluded it is unlikely they removed
seeds (Mattos and Orrock 2010). The December
presentation showed the most pronounced effect of
allowing small mammal granivores access, potentially
due to dwindling food resources (Haught and Myster
2008). However, not all seeds removed by small
mammals were likely consumed, as all of the granivorous rodents known on Fort Bragg, except S. hispidus
(Cameron and Spencer 1981), S. palustris (Chapman
and Willner 1981), and Reithrodontomys humulis
(Stalling 1997), exhibit flexible feeding patterns with
scatterhoarding or caching traits.
The observed differences in removal suggest that
habitat type and fire may jointly influence the spatial
distribution of the seeds of woody exotic species
within fire-dependent longleaf pine savannas. Yet the
implications of these findings for the establishment
and spread of woody exotic species remain unclear
because we did not follow the seeds post-removal.
Granivores could have consumed, dispersed or cached
seeds following removal (Vander Wall et al. 2005).
Moreover, seeds in sites unsuitable for germination are
likely never to establish, and therefore their removal is
unlikely to affect plant recruitment (Maron and Simms
1997). When populations are limited by seeds rather
than microsite availability, however, there is consistent evidence suggesting that post-dispersal seed
removal negatively affect woody plant establishment
and demography by decreasing the number of viable
seeds that are available to establish in suitable sites
(Ostfeld et al. 1997; Santos and Telleria 1997; Orrock
et al. 2006; Nunez et al. 2008; Zwolak et al. 2010). In
the longleaf pine ecosystem, each of the focal species
is known to invade xeric upland (Kush and Meldahl
2000; Provencher et al. 2000; Renne et al. 2000;
Stocker and Hupp 2008) and mesic ecotone plant
communities (Herring and Judd 1995; Drew et al.
1998; Renne et al. 2000), which suggests that seeds of
all four species have the potential to germinate in both
upland and ecotone habitats if they escape postdispersal predation. Despite this, invasion of our study
system by the focal species remains rare. This is
evidence that the focal species are seed limited,
despite the large number of seeds a single tree can
produce. Given these findings and the observed
differences in seed removal patterns, we conclude
that granivores are limiting exotic woody invasions in
ecotone habitat, but hypothesize that the influence of
granivores in ecotone habitat is diminished during the
first growing season post-fire. By corollary, and
assuming that these findings hold for the duration of
the winter season, there may be predictable sites
offering escape from seed predation in this ecosystem.
Studies evaluating establishment limitation should be
conducted to assess invasion vulnerability in such
areas.
Generalizations about the ultimate role of any
process in generating pattern are limited by our
knowledge about how the process varies in space
and time. Biological invasions are inherently a
spatially and temporally dynamic phenomenon. With
regard to invasive plants, post-dispersal establishment
is affected by multiple biotic and abiotic factors. An
understanding of the patterns of post-dispersal seed
removal may be useful for managing natural savanna
ecosystems that are under the threat of encroachment
from exotic woody species, and for increasing our
understanding of invasion dynamics.
Acknowledgments We thank B. Mattingly and J. Orrock for
assisting with study design and T.J. Benson for help with data
analysis. This research was supported by the Department of
Defense USACE—ERDC—Construction Engineering Research
Laboratory. J. Gray provided invaluable logistical support and
assistance in the field. We also thank A. Davis and J. Dalling for
providing helpful comments on early versions of the manuscript.
References
Alcantara JM, Rey PJ, Sanchez-Lafuente AM, Valera F (2000)
Early effects of rodent post-dispersal seed predation on the
outcome of the plant-seed disperser interaction. Oikos
88:362–370
Auld TD, Denham AJ (2001) The impact of seed predation by
mammals on post-fire seed accumulation in the endangered
shrub Grevillea caleyi (Proteaceae). Biol Conserv
97:377–385
Beckage B, Platt WJ, Gross LJ (2009) Vegetation, fire and
feedbacks: a disturbance-mediated model of savannas. Am
Nat 174:805–818
Bolker MB, Brooks ME, Clark CJ, Geange SW, Poulsen JR,
Stevens MHH, White JS (2009) Generalized linear mixed
models: a practical guide for ecology and evolution. Trends
Ecol Evol 24:127–135
123
J. S. Krall et al.
Boman JS, Casper BB (1995) Differential postdispersal seed
predation in disturbed and intact temperate forest. Am Midl
Nat 134:107–116
Bond WJ, Keeley JE (2005) Fire as a global ‘herbivore’: the
ecology and evolution of flammable ecosystems. Trends
Ecol Evol 20:387–394
Bowers MA (1990) Exploitation of seed aggregates by Merriam’s kangaroo rat: harvesting rates and predatory risk.
Ecology 71:2334–2344
Bowers MA (1993) Influence of herbivorous mammals on an
old-field plant community: years 1–4 after disturbance.
Oikos 67:129–141
Brooks ML, D’Antonio CM, Richardson DM, Grace JB, Keeley
JE, DiTomaso JM, Hobbs RJ, Pellant M, Pyke D (2004)
Effects of invasive alien plants on fire regimes. Bioscience
54:677–688
Brown JH, Heske EJ (1990) Control of a desert-grassland transition by a keystone rodent guild. Science 250:1705–1707
Cameron GN, Spencer SR (1981) Sigmodon hispidus. Mammalian Species 158:1–9
Carrillo-Gavilan MA, Lalague H, Vila M (2010) Comparing
seed removal of 16 pine species differing in invasiveness.
Biol Invasions 12:2233–2242
Chapman JA, Willner GR (1981) Sylvilagus palustris. Mammalian Species 153:1–3
Clark CJ, Poulsen JR, Levey DJ, Osenberg CW (2007) Are plant
populations seed limited? A critique and meta-analysis of
seed addition experiments. Am Nat 170:128–142
Côté M, Ferron J, Gagnon R (2005) Invertebrate predation of
postdispersal seeds and juvenile seedlings of black spruce
(Picea mariana) in the boreal forest of eastern Canada. Can
J For Res 35:674–681
Craig MT, Orrock JL, Brudvig LA (2011) Edge-mediated patterns of seed removal in experimentally connected and
fragmented landscapes. Landscape Ecol 26:1373–1381
Crist TO, Wiens JA (1994) Scale effects of vegetation on forager
movement and seed harvesting by ants. Oikos 69:37–46
Drew MB, Kirkman LK, Gholson AK (1998) The vascular flora
of Ichauway, Baker County, Georgia: a remnant longleaf
pine/wiregrass ecosystem. Castanea 63:1–24
Drewa PB, Platt WJ, Moser EB (2002) Fire effects on resprouting of shrubs in headwaters of southeastern longleaf pine
savannas. Ecology 83:755–767
Duncan WH, Piercy PL, Starling RJ (1953) Toxicological
studies of southeastern plants. I. Leguminosae. Econ Bot
9:243–255
Ellair DP, Platt WJ (2012) Fuel composition influences fire
characteristics and understory hardwoods in pine savanna.
J Ecol 101:192–201
Fewell JH (1988) Energetic and time costs of foraging in harvester ants, Pogonomyrmex occidentalis. Behav Ecol Sociobiol 22:401–408
Garcia-Castano JL, Kollmann J, Jordano P (2006) Spatial variation of post-dispersal seed removal by rodents in highland microhabitats of Spain and Switzerland. Seed Sci Res
16:213–222
Gibb H, Parr CL (2010) How does habitat complexity affect ant
foraging success? A test using functional measures on three
continents. Oecologia 164:1061–1073
Golley FB, Gentry JB (1964) Bioenergetics of the southern
harvester ant, Pogonomyrmex badius. Ecology 45:217–225
123
Grady JM, Hoffmann WA (2012) Caught in a fire trap: recurring
fire creates stable size equilibria in woody resprouters.
Ecology 93:2052–2060
Grant G, More LJ, McKenzie NH, Dorward PM, Buchan WC,
Telek L, Pusztai A (1995) Nutritional and haemagglutination properties of several tropical seeds. J Agric Sci
124:437–445
Haught JE, Myster RW (2008) Effects of species, density, season and prairie-type on post-dispersal seed removal in
Oklahoma. Am Midl Nat 159:482–488
Herring BJ, Judd WS (1995) A floristic study of Ichetucknee
Springs State Park, Suwannee and Columbia counties,
Florida. Castanea 60:318–369
Hiers JK, Mitchell RJ (2007) The influence of burning and light
availability on N2-fixation of native legumes in longleaf
pine woodlands. J Torrey Bot Soc 134:398–409
Hulme PE (1994) Post-dispersal seed predation in grassland: its
magnitude and sources of variation. J Ecol 82:645–652
Hulme PE (1998) Post-dispersal seed predation: consequences
for plant demography and evolution. Perspect Plant Ecol
Evol Syst 1:32–46
Jacob J, Brown JS (2000) Microhabitat use, giving-up densities
and temporal activity as short- and long-term anti-predator
behaviors in common voles. Oikos 91:131–138
Janzen DH (1971) Seed predation by animals. Annu Rev Ecol
Syst 2:465–492
Johnson JB, Omland KS (2004) Model selection in ecology and
evolution. Trends Ecol Evol 19:101–108
Kelrick MI, MacMahon JA, Parmenter RR, Sisson DV (1986)
Native seed preferences of shrub-steppe rodents, birds and
ants: the relationship of seed attributes and seed use.
Oecologica 68:327–337
Kirkman LK, Goebel PC, Palik BJ, West LT (2004) Predicting
plant species diversity in a longleaf pine landscape. Ecoscience 11:80–93
Kollmann J, Coomes DA, White SM (1998) Consistencies in
post-dispersal seed predation of temperate fleshy-fruited
species among seasons, years and sites. Funct Ecol
12:683–690
Kotler BP (1984) Risk of predation and the structure of desert
rodent communities. Ecology 65:689–701
Krefting LW, Ahlgren CE (1974) Small vertebrates and vegetation changes after fire in a mixed conifer-hardwood forest. Ecology 55:1391–1398
Kuppinger DM, Jenkins MA, White PS (2010) Predicting the postfire establishment and persistence of an invasive species
across a complex landscape. Biol Invasions 12:3473–3484
Kush JS, Meldahl RS (2000) Composition of a virgin stand of
longleaf pine in south Alabama. Castanea 65:56–63
Lackey JA, Huckaby DG, Ormiston BG (1985) Peromyscus
leucopus. Mammalian Species 247:1–10
Leishman MR, Wright IJ, Moles AT, Westoby M (2000) The
evolutionary ecology of seed size. In: Fenner M (ed) Seeds:
the ecology of regeneration in plant communities. CAB
International, Wallingford, pp 31–57
Lonsdale WM, Miller IL (1993) Fire as a management tool for
the tropical woody weed: Mimosa pigra in northern Australia. J Environ Manage 39:77–87
Lundgren JG, Rosentrater KA (2007) The strength of seeds and
their destruction by granivorous insects. Arthropod Plant
Interact 1:93–99
Exotic seed removal and fire in longleaf pine savanna
MacMahon JA, Mull JF, Crist TO (2000) Harvester ants
(Pogonomyrmex spp.): their community and ecosystem
influences. Annu Rev Ecol Syst 31:265–291
Mandle L, Bufford JL, Schmidt IB, Daehler CC (2011) Woody exotic
plant invasions and fire: reciprocal impacts and consequences
for native ecosystems. Biol Invasions 13:1815–1827
Manson RH, Stiles EW (1998) Links between microhabitat
preferences and seed predation by small vertebrates in old
fields. Oikos 82:37–50
Maron JL, Simms EL (1997) Effect of seed predation on seed
bank size and seedling recruitment of bush lupine (Lupinus
arboreus). Oecologia 111:76–83
Matias L, Mendoza I, Zamora R (2009) Consistent pattern of
habitat and species selection by post-dispersal seed predators in a Mediterranean mosaic landscape. Plant Ecol
203:137–147
Mattos KJ, Orrock JL (2010) Behavioral consequences of plant
invasion: an invasive plant alters rodent antipredator
behavior. Behav Ecol 21:556–561
Mazia CN, Chaneton EJ, Machera M, Uchitel A, Feler MV,
Ghersa CM (2010) Antagonistic effects of large- and
small-scale disturbances on exotic tree invasion in a native
tussock grassland relict. Biol Invasions 12:3109–3122
McCoy ED, Kaiser BW (1990) Changes in foraging activity of
the southern harvester ant Pogonomyrmex badius (Latreille) in response to fire. Am Midl Nat 123:112–123
Meyer R (2010) Albizia julibrissin. In: Fire effects information
system, [Online]. U.S. Department of Agriculture, Forest
Service, Rocky Mountain Research Station, Fire Sciences
Laboratory (Producer). Available: http://www.fs.fed.us/
database/feis/. 20 Jan 2013
Meyer R (2011) Triadica sebifera. In: Fire effects information
system, [Online]. U.S. Department of Agriculture, Forest
Service, Rocky Mountain Research Station, Fire Sciences
Laboratory (Producer). Available: http://www.fs.fed.us/
database/feis/. 20 Jan 2013
Mittelbach GG, Gross KL (1984) Experimental studies of seed
predation in old-fields. Oecologia 65:7–13
Munger GT (2003) Elaeagnus umbellata. In: Fire effects
information system, [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station,
Fire Sciences Laboratory (Producer). Available: http://
www.fs.fed.us/database/feis/. 20 Jan 2013
Nathan R, Muller-Landau HC (2000) Spatial patterns of seed
dispersal, their determinants and consequences for
recruitment. Trends Ecol Evol 15:278–285
Ne’eman G, Goubitz S, Nathan R (2004) Reproductive traits of
Pinus halepensis in the light of fire—a critical review.
Plant Ecol 171:69–79
Niehaus AC, Wilson RS, Storm JJ, Angilletta MJ (2012) Fall
field crickets did not acclimate to simulated seasonal
changes in temperature. J Comp Physiol B 182:199–207
Nunez MA, Simberloff D, Relva MA (2008) Seed predation as a
barrier to alien conifer invasions. Biol Invasions 10:1389–1398
Oelrichs PB, Hill MW, Vallely PJ, MacLaod JK, Molinski TF
(1983) Toxic tetranortriterpenes of the fruit of Melia azedarach. Phytochemistry 22:531–534
Olson MS, Platt WJ (1995) Effects of habitat and growing
season on fires on resprouting of shrubs in longleaf pine
savannas. Vegetatio 119:101–118
Onofri A, Carbonell EA, Piepho HP, Mortimer AM, Cousens
RD (2010) Current statistical issues in Weed Research.
Weed Res 50:5–24
Orrock JL, Damschen EI (2005) Corridors cause differential
seed predation. Ecol Appl 15:793–798
Orrock JL, Danielson BJ, Brinkerhoff RJ (2004) Rodent foraging is affected by indirect, but not by direct, cues of
predation risk. Behav Ecol 15:433–437
Orrock JL, Levey DJ, Danielson BJ, Damschen EI (2006) Seed
predation, not seed dispersal, explains the landscape-level
abundance of an early-successional plant. Ecology
94:838–845
Ostfeld RS, Manson RH, Canham CD (1997) Effects of rodents
on survival of tree seeds and seedlings invading old fields.
Ecology 78:1531–1542
Pons J, Pausas JG (2007) Rodent acorn selection in a Mediterranean oak landscape. Ecol Res 22:535–541
Provencher L, Herring BJ, Gordon DR, Rodgers HL, Tanner
GW, Brennan LA, Hardesty JL (2000) Restoration of
northwest Florida sandhills through harvest of Pinus
clausa. Restor Ecol 8:175–185
Rebertus AJ, Williamson GB, Moser EB (1989) Longleaf pine
pyrogenicity and turkey oak mortality in Florida xeric
sandhills. Ecology 70:60–70
Reed AW, Kaufman GA, Kaufman DW (2004) Influence of fire,
topography, and consumer abundance on seed predation in
tallgrass prairie. Can J Zool 82:1459–1467
Reed AW, Kaufman GA, Kaufman DW (2005) Rodent seed
predation and GUDs: effect of burning and topography.
Can J Zool 83:1279–1285
Reed AW, Kaufman GA, Kaufman DW (2006) Effect of plant
litter on seed predation in three prairie types. Am Midl Nat
155:278–285
Renne IJ, Gauthreaux SA, Gresham CA (2000) Seed dispersal of
the Chinese tallow tree (Sapium sebiferum (L.) Roxb.) by
birds in coastal South Carolina. Am Midl Nat 144:202–215
Rey PJ, Garrido JL, Alcantara JM, Ramirez JM, Aguilera A,
Garcia L, Manzaneda AJ, Fernandez R (2002) Spatial
variation in ant and rodent post-dispersal predation of
vertebrate-dispersed seeds. Funct Ecol 16:773–781
Romero C, Bolker BM, Edwards CE (2009) Stem responses to
damage: the evolutionary ecology of Quercus species in
contrasting fire regimes. New Phytol 182:261–271
Santos T, Telleria JL (1997) Vertebrate predation on Holm oak,
Quercus ilex, acorns in a fragmented habitat: effects on
seedling recruitment. For Ecol Manage 98:181–187
Satterthwaite WH (2007) The importance of dispersal in
determining seed versus safe site limitation of plant populations. Plant Ecol 193:113–130
Schnurr JL, Canham CD, Ostfeld RS, Inouye RS (2004)
Neighborhood analysis of small-mammal dynamics:
impacts of seed predation and seedling establishment.
Ecology 85:741–755
Sorrie BA, Gray JB, Crutchfield PJ (2006) The vascular flora of
the longleaf pine ecosystem of Fort Bragg and Weymouth
Woods, North Carolina. Castanea 71:129–161
Sousa WP (1984) The role of disturbance in natural communities. Annu Rev Ecol Syst 15:353–391
Stalling DT (1997) Reithrodontomys humulis. Mammalian
Species 565:1–6
123
J. S. Krall et al.
Stevens JT, Beckage B (2010) Fire effects on demography of the
invasive shrub Brazilian pepper (Schinus terebinthifolius)
in Florida pine savannas. Nat Areas J 30:53–63
Stocker R, Hupp KVS (2008) Chapter 6: fire and nonnative
invasive plants in the southeast bioregion. USDA For Serv
Gen Tech Rep 6:91–112
Taylor CM, Hastings A (2005) Allee effects in biological
invasions. Ecol Lett 8:895–908
Terigar BG, Balasubramanian S, Boldor D (2010) Effect of
storage conditions on the oil quality of Chinese tallow tree
seeds. J Am Oil Chem Soc 87:573–582
Tester JP (1965) Effects of a controlled burn on small vertebrates in a Minnesota oak savanna. Am Midl Nat
74:240–243
Thaxton JM, Platt WJ (2006) Small-scale fuel variation alters
fire intensity and shrub abundance in a pine savanna.
Ecology 87:1331–1337
Theoharides KA, Dukes JS (2007) Plant invasion across space
and time: factors affecting nonindigenous species success
during four stages of invasion. New Phytol 176:256–273
Thompson K, Rabinowitz D (1989) Do big plants have big
seeds? Am Nat 133:722–728
Todorović S, Božić D, Simonović A, Filipović B, Dragićević M,
Giba Z, Grubišić D (2010) Interaction of fire-related cues in
seed germination of the potentially invasive species Paulownia tomentosa Steud. Plant Species Biol 25:193–202
123
Vander Wall SB, Kuhn KM, Beck MJ (2005) Seed removal,
seed predation, and secondary dispersal. Ecology
86:801–806
Vaz Ferreira A, Bruna EM, Vasconcelos HL (2011) Seed predators limit plant recruitment in Neotropical savannas.
Oikos 120:1013–1022
Waggy MA (2009) Melia azedarach. In: Fire effects information system, [Online]. U.S. Department of Agriculture,
Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available: http://www.fs.
fed.us/database/feis/. 20 Jan 2013
Weakley AS (2012) Flora of the southern and mid-Atlantic
states. University of North Carolina at Chapel Hill, North
Carolina Botanical Garden, University of North Carolina
Herbarium (Producer). Available: http://www.herbarium.
unc.edu/flora.htm. 20 Jan 2013
Whelan CJ, Willson MF, Tuma CA, Souza-Pinto I (1991)
Spatial and temporal patterns of postdispersal seed predation. Can J Bot 69:428–436
Zwolak R, Pearson DE, Ortega YK, Crone EE (2010) Fire and
mice: seed predation moderates fire’s influence on conifer
recruitment. Ecology 91:1124–1131
Zwolak R, Pearson DE, Ortega YK, Crone EE (2012) Mechanisms driving postfire abundance of a generalist mammal.
Can J Zool 90:51–60